Assessment of the ability to process semantic and phonological aspects of words in aphasia: A multi-measurement approach

Background: Aphasia can disrupt processing of semantic and/or phonological aspects of words and each of these domains involves multiple operations. Adequate assessment of word processing requires multiple measurements probing in each domain. Aims: This paper aims to facilitate accurate and comprehensive testing of the multiple operations involved in semantic and phonological processing of spoken words. Methods & Procedures: Normative data (means and standard deviations) from aphasic samples on six measures of input semantic processing and five measures of input phonological processing are provided. Additionally, scores on these measures are correlated with performance on a picture-naming test to determine whether they could predict performance in word production. The benefits of using a multi-measurement approach are discussed. Outcome & Results: We demonstrate how these norms can be used to establish z-score profiles of an individual's semantic and phonological abilities with reference to the aphasic sample. Additionally, we provide evidence that the measures of input processing on the semantic and phonological batteries can predict response patterns in a production task, picture naming. z-scores on the measures were positively correlated with picture-naming accuracy, and measures in each domain followed a predicted pattern of correlation with error patterns in naming. Conclusions: Using the means and standard deviations from this study, a clinician or researcher can establish z-score profiles of participants that will indicate the severity of semantic or phonological impairment compared with other individuals with aphasia. Comparison with this reference group will aid in the development of a therapy programme that is at an appropriate level of difficulty and will facilitate interpretation of progress being made in treatment. Additionally, this battery of input measures can be used to estimate patterns of output (correct and erroneous responses) in picture naming.